Lindsay Edwards
Duke University, NC·Medicine & Clinical Research·Early-career
Cardiomyopathy, a weakening of the heart muscle, is a life-threatening and fairly common late effect of childhood cancer treatment. More than half of pediatric cancer survivors received heart-damaging therapies, so they get a heart ultrasound (echocardiogram) every 2 to 5 years.
- Fields
Medicine & Clinical Research
Computer Science
Biomedical Engineering
- Topics
cardiomyopathy
childhood cancer survivors
pediatric cancer
- Grants
- 1 known grant, through Jul 2031
What they work on
- Optimize a machine learning pipeline that predicts cardiomyopathy in pediatric cancer survivors from routinely collected echocardiographic and clinical data.
- Improve prediction using synthetic data generation, which creates realistic artificial patient records for training, and multimodal fusion modeling, which combines heart-imaging measurements with clinical information.
- Run focus groups with pediatric cardiologists and oncologists, the intended clinical users, so their views shape how the prediction model is built and how its interface presents results.
More about the research
Specialized core labs can spot subtle changes up to 4 years before the disease is diagnosed. In routine clinics, however, readers vary too much for these measurements to predict who will get sick. By the time the heart's pumping is clearly impaired, the damage is likely permanent. In other patients, starting heart-protective therapy early can delay heart failure, so predicting which survivors are at risk could allow closer monitoring and earlier treatment.
Why it matters
Childhood cancer survivors who received heart-damaging treatment face a lasting risk of cardiomyopathy and heart failure. A model that flags at-risk survivors from routine echocardiograms and clinical records could let doctors monitor them more closely. It could also let them start protective therapy before the damage becomes permanent, and could become the basis for clinical decision support tools.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Oguz Akbilgic
Wake Forest University, NC·Medicine & Clinical Research
Develop an AI model that predicts future heart failure with preserved ejection fraction (HFpEF) from ECG data, using carefully verified data from NIH-funded studies.
Weidan Cao
Ohio State University, OH·Medicine & Clinical Research
Design and develop, with stakeholders, a human-centered smart system that pairs wearable sensors with a conversational large language model chatbot to collect health data at home with little patient effort.
Satoru Kobayashi
New York Institute of Technology, NY·Pharmacology
Measure how lysosomal membrane permeabilization (LMP) contributes to doxorubicin-induced heart injury, using an LMP reporter, the Lyso-tag lysosome isolation system and super-resolution imaging in heart cells and mice.
The school
Duke University: 61 labs on PI Pages, $1.6B a year on research, 417 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.